Partial Differential Equations for Zooming, Deinterlacing and Dejittering

  • Authors:
  • Frank Lenzen;Otmar Scherzer

  • Affiliations:
  • Heidelberg Collaboration for Image Processing, University of Heidelberg, Heidelberg, Germany 69115;Computational Science Center, University of Vienna, Vienna, Austria 1090 and Johann Radon Institute for Computational and Applied Mathematics (RICAM), Austrian Academy of Sciences, Linz, Austria 4 ...

  • Venue:
  • International Journal of Computer Vision
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, for imaging applications, we introduce partial differential equations (PDEs), which allow for correcting displacement errors, for dejittering, and for deinterlacing, respectively, in multi-channel data. These equations are derived via semi-groups for non-convex energy functionals. As a particular example, for gray valued data, we find the mean curvature equation and the corresponding non-convex energy functional. As a further application for correction of displacement errors we study image interpolation, in particular zooming, of digital color images. For actual image zooming, the solutions of the proposed PDEs are projected onto a space of functions satisfying interpolation constraints. A comparison of the test results with standard and state-of-the-art interpolation algorithms shows the competitiveness of this approach.